import requests
import json
import time
import re
import datetime
import yaml
import pandas as pd
import numpy as np
import warnings
warnings.filterwarnings('ignore')
import plotly.graph_objects as go
import seaborn as sns
import matplotlib.pyplot as plt

def eda(data):
    output=[]
    for col in data.columns:
        duplicatedvalue = data[col].duplicated().sum()
        duplicatedrows = data.duplicated().sum()
        missingvalue = np.sum(pd.isna(data[col]))
        uniquevalue = data[col].nunique()
        datatype = str(data[col].dtype)
        
        output.append([col, duplicatedvalue, duplicatedrows, missingvalue, uniquevalue, datatype])
        
    output = pd.DataFrame(output) 
    output.columns = ['Features', 'Duplicated Values', 'Duplicated Rows', 'Missing Values', 'Unique Values', 'Data Type']
    display(output)

def perc_on_bar(plot, feature):
    total = len(feature)
    for p in ax.patches:
        percentage = "{:.1f}%".format(100 * p.get_height() / total)
        x = p.get_x() + p.get_width() / 2 - 0.05
        y = p.get_y() + p.get_height()
        ax.annotate(percentage, (x, y), size=12)
    plt.show()

with open ('config.yml') as ymlfile:
    cfg = yaml.safe_load(ymlfile)
    oanda_api_key = cfg['creds']['oanda_api']

provider_authorization = 'Bearer {0}'.format(oanda_api_key)

headers = {
    'Content-Type': 'application/json',
    'Authorization': provider_authorization,
}

Daily Candlesticks

params_D = (
    ('price', 'BA'),
    ('from', '2016-10-17T15:00:00.000000000Z'),
    ('granularity', 'D'),
)

H4 Candlesticks

params_4H = (
    ('price', 'BA'),
    ('count', '5000'),
    ('granularity', 'H4'),
)

Get the Candlestick Data

response = requests.get('https://api-fxpractice.oanda.com/v3/instruments/EUR_USD/candles', 
                        headers=headers, 
                        params=params_4H).json()

response.keys()response['candles'][0]

Save as CSV

output = []
all_candlesticks = response['candles']

for i in range (len(all_candlesticks)):
    output.append([response['candles'][i]['time'], 
                  response['candles'][i]['volume'], 
                  response['candles'][i]['bid']['o'],
                  response['candles'][i]['bid']['h'],
                  response['candles'][i]['bid']['l'],
                  response['candles'][i]['bid']['c']])

output = pd.DataFrame(output)
output.columns = ['Date', 'Volume', 'Open', 'High', 'Low', 'Close']

eurusd_data = output.to_csv('EURUSD_4H.csv', header = True, index = False)

eurusd_data = pd.read_csv('EURUSD_4H.csv')

eurusd_data.shape

(5000, 6)

eurusd_data

	Date	Volume	Open	High	Low	Close
0	2018-06-04T17:00:00.000000000Z	4014	1.16852	1.17016	1.16848	1.16956
1	2018-06-04T21:00:00.000000000Z	4032	1.16939	1.17029	1.16901	1.16930
2	2018-06-05T01:00:00.000000000Z	3802	1.16928	1.16968	1.16853	1.16865
3	2018-06-05T05:00:00.000000000Z	12695	1.16862	1.17162	1.16825	1.17039
4	2018-06-05T09:00:00.000000000Z	13544	1.17034	1.17058	1.16615	1.16657
...	...	...	...	...	...	...
4995	2021-08-20T01:00:00.000000000Z	5024	1.16814	1.16869	1.16751	1.16853
4996	2021-08-20T05:00:00.000000000Z	15320	1.16851	1.16886	1.16701	1.16734
4997	2021-08-20T09:00:00.000000000Z	13877	1.16733	1.16828	1.16643	1.16678
4998	2021-08-20T13:00:00.000000000Z	23033	1.16676	1.17039	1.16634	1.17026
4999	2021-08-20T17:00:00.000000000Z	6450	1.17024	1.17039	1.16915	1.16981

5000 rows × 6 columns

fig = go.Figure(data=[go.Candlestick(x=eurusd_data['Date'],
                open=eurusd_data['Open'],
                high=eurusd_data['Close'],
                low=eurusd_data['Low'],
                close=eurusd_data['Close'])])

fig.show()

Feature Engineering

CandleStick Data

eurusd_data['O-H'] = eurusd_data['Open'] - eurusd_data['High']
eurusd_data['O-L'] = eurusd_data['Open'] - eurusd_data['Low']
eurusd_data['O-C'] = eurusd_data['Open'] - eurusd_data['Close']
eurusd_data['H-L'] = eurusd_data['High'] - eurusd_data['Low']
eurusd_data['H-C'] = eurusd_data['High'] - eurusd_data['Close']
eurusd_data['L-C'] = eurusd_data['Low'] - eurusd_data['Close']

eurusd_data.head()

	Date	Volume	Open	High	Low	Close	O-H	O-L	O-C	H-L	H-C	L-C
0	2018-06-04T17:00:00.000000000Z	4014	1.16852	1.17016	1.16848	1.16956	-0.00164	0.00004	-0.00104	0.00168	0.00060	-0.00108
1	2018-06-04T21:00:00.000000000Z	4032	1.16939	1.17029	1.16901	1.16930	-0.00090	0.00038	0.00009	0.00128	0.00099	-0.00029
2	2018-06-05T01:00:00.000000000Z	3802	1.16928	1.16968	1.16853	1.16865	-0.00040	0.00075	0.00063	0.00115	0.00103	-0.00012
3	2018-06-05T05:00:00.000000000Z	12695	1.16862	1.17162	1.16825	1.17039	-0.00300	0.00037	-0.00177	0.00337	0.00123	-0.00214
4	2018-06-05T09:00:00.000000000Z	13544	1.17034	1.17058	1.16615	1.16657	-0.00024	0.00419	0.00377	0.00443	0.00401	-0.00042

eurusd_data['Direction'] = eurusd_data['O-C'].apply(lambda x: 'UP' if x<0 else 'DOWN')

eurusd_data.tail(25)

	Date	Volume	Open	High	Low	Close	O-H	O-L	O-C	H-L	H-C	L-C	Direction
4975	2021-08-16T17:00:00.000000000Z	5151	1.17747	1.17789	1.17738	1.17755	-0.00042	0.00009	-0.00008	0.00051	0.00034	-0.00017	UP
4976	2021-08-16T21:00:00.000000000Z	2349	1.17764	1.17797	1.17696	1.17733	-0.00033	0.00068	0.00031	0.00101	0.00064	-0.00037	DOWN
4977	2021-08-17T01:00:00.000000000Z	6649	1.17731	1.17761	1.17662	1.17679	-0.00030	0.00069	0.00052	0.00099	0.00082	-0.00017	DOWN
4978	2021-08-17T05:00:00.000000000Z	20223	1.17680	1.17843	1.17596	1.17767	-0.00163	0.00084	-0.00087	0.00247	0.00076	-0.00171	UP
4979	2021-08-17T09:00:00.000000000Z	21457	1.17768	1.17770	1.17406	1.17435	-0.00002	0.00362	0.00333	0.00364	0.00335	-0.00029	DOWN
4980	2021-08-17T13:00:00.000000000Z	26862	1.17434	1.17442	1.17079	1.17081	-0.00008	0.00355	0.00353	0.00363	0.00361	-0.00002	DOWN
4981	2021-08-17T17:00:00.000000000Z	9874	1.17079	1.17192	1.17072	1.17078	-0.00113	0.00007	0.00001	0.00120	0.00114	-0.00006	DOWN
4982	2021-08-17T21:00:00.000000000Z	3548	1.17093	1.17128	1.17013	1.17115	-0.00035	0.00080	-0.00022	0.00115	0.00013	-0.00102	UP
4983	2021-08-18T01:00:00.000000000Z	6521	1.17114	1.17224	1.17094	1.17177	-0.00110	0.00020	-0.00063	0.00130	0.00047	-0.00083	UP
4984	2021-08-18T05:00:00.000000000Z	12408	1.17176	1.17291	1.17120	1.17196	-0.00115	0.00056	-0.00020	0.00171	0.00095	-0.00076	UP
4985	2021-08-18T09:00:00.000000000Z	16056	1.17197	1.17256	1.17053	1.17171	-0.00059	0.00144	0.00026	0.00203	0.00085	-0.00118	DOWN
4986	2021-08-18T13:00:00.000000000Z	21810	1.17173	1.17206	1.16929	1.16987	-0.00033	0.00244	0.00186	0.00277	0.00219	-0.00058	DOWN
4987	2021-08-18T17:00:00.000000000Z	24450	1.16988	1.17417	1.16973	1.17075	-0.00429	0.00015	-0.00087	0.00444	0.00342	-0.00102	UP
4988	2021-08-18T21:00:00.000000000Z	4977	1.17073	1.17150	1.16833	1.16868	-0.00077	0.00240	0.00205	0.00317	0.00282	-0.00035	DOWN
4989	2021-08-19T01:00:00.000000000Z	9878	1.16870	1.16930	1.16648	1.16763	-0.00060	0.00222	0.00107	0.00282	0.00167	-0.00115	DOWN
4990	2021-08-19T05:00:00.000000000Z	25661	1.16761	1.16966	1.16682	1.16862	-0.00205	0.00079	-0.00101	0.00284	0.00104	-0.00180	UP
4991	2021-08-19T09:00:00.000000000Z	20884	1.16863	1.17036	1.16795	1.16981	-0.00173	0.00068	-0.00118	0.00241	0.00055	-0.00186	UP
4992	2021-08-19T13:00:00.000000000Z	26254	1.16980	1.17019	1.16782	1.16877	-0.00039	0.00198	0.00103	0.00237	0.00142	-0.00095	DOWN
4993	2021-08-19T17:00:00.000000000Z	9181	1.16878	1.16879	1.16719	1.16755	-0.00001	0.00159	0.00123	0.00160	0.00124	-0.00036	DOWN
4994	2021-08-19T21:00:00.000000000Z	1958	1.16755	1.16832	1.16745	1.16816	-0.00077	0.00010	-0.00061	0.00087	0.00016	-0.00071	UP
4995	2021-08-20T01:00:00.000000000Z	5024	1.16814	1.16869	1.16751	1.16853	-0.00055	0.00063	-0.00039	0.00118	0.00016	-0.00102	UP
4996	2021-08-20T05:00:00.000000000Z	15320	1.16851	1.16886	1.16701	1.16734	-0.00035	0.00150	0.00117	0.00185	0.00152	-0.00033	DOWN
4997	2021-08-20T09:00:00.000000000Z	13877	1.16733	1.16828	1.16643	1.16678	-0.00095	0.00090	0.00055	0.00185	0.00150	-0.00035	DOWN
4998	2021-08-20T13:00:00.000000000Z	23033	1.16676	1.17039	1.16634	1.17026	-0.00363	0.00042	-0.00350	0.00405	0.00013	-0.00392	UP
4999	2021-08-20T17:00:00.000000000Z	6450	1.17024	1.17039	1.16915	1.16981	-0.00015	0.00109	0.00043	0.00124	0.00058	-0.00066	DOWN

eda (eurusd_data)

	Features	Duplicated Values	Duplicated Rows	Missing Values	Unique Values	Data Type
0	Date	0	0	0	5000	object
1	Volume	608	0	0	4392	int64
2	Open	883	0	0	4117	float64
3	High	865	0	0	4135	float64
4	Low	878	0	0	4122	float64
5	Close	857	0	0	4143	float64
6	O-H	3977	0	0	1023	float64
7	O-L	3992	0	0	1008	float64
8	O-C	3447	0	0	1553	float64
9	H-L	3689	0	0	1311	float64
10	H-C	3968	0	0	1032	float64
11	L-C	4003	0	0	997	float64
12	Direction	4998	0	0	2	object

plt.figure(figsize=(20,10))
plt.xticks(rotation = 90)
ax = sns.countplot(eurusd_data['Direction'].tail(cc), order=eurusd_data['Direction'].tail(cc).value_counts().index)
perc_on_bar(ax, eurusd_data['Direction'].tail(cc))

Insights

• The trend is bearish

cc = 90 #candlestick count
fig = go.Figure(data=[go.Candlestick(x=eurusd_data['Date'].tail(cc),
                open=eurusd_data['Open'].tail(cc),
                high=eurusd_data['Close'].tail(cc),
                low=eurusd_data['Low'].tail(cc),
                close=eurusd_data['Close'].tail(cc))])

fig.update_layout(xaxis_rangeslider_visible=False)
fig.show()

Date

Current: 2021-08-20T17:00:00.000000000Z Expected: 2021/8/20, 17:00, Friday